One type of search performed by a user with a search engine is a location related search, for example, searching for information about a traffic condition near a certain location, searching for information about a distribution of convenience stores near a certain location, searching for information about a distribution of bus stations near a certain location, etc. A location can be represented by coordinates or by location name. Since most existing search engines are text based, location name is often used to represent a location. According to a general solution, a user enters keywords in an interface provided by a search engine. The keywords include a location name such as “Summer Palace”; the keywords also include an information type to be learned, such as “traffic condition”. The search engine searches content containing strings “Summer Palace” and “traffic condition” and returns the content back to the user as results. Some search engine may perform extension on keywords and take the extended keywords as keywords that are actually used. For example, according to one type of literal extension, “traffic condition” may be extended as “traffic” and “condition”, so that not only content containing string “traffic condition” is acquired, but also content only containing string “traffic” or “condition” are also acquired. According to one type of semantic extension, for example, “traffic condition” is extended into “traffic situation”, “traffic problem” etc. Literal extension and semantic extension may be combined together so as to return more search results to the user. These solutions all take a location name as a text attribute. However, a location is also a kind of geographical information. There is a need for a solution to use geographical information in a search related to locations.